Robust Biomolecular Complex Design Across Protein Conformational Landscapes

Abstract

Proteins populate conformational ensembles, yet structure-based biomolecular design typically optimizes candidates against a single target conformation. Consequently, a candidate that fits one state can lose favorable interactions or develop steric clashes when the target adopts another. We introduce FlexEvo, a model-agnostic evolutionary framework that adapts candidates once at inference time from a single target conformation to improve compatibility with alternative natural conformations unseen during adaptation, without retraining the source model or requiring a conformational ensemble. FlexEvo casts cross-state adaptation as geometry-constrained bi-objective optimization, balancing preservation of input-state interactions against robustness to plausible conformational perturbations. To limit the search space and reduce invalid structural edits, geometry-derived FlexBoxes define protected anchor regions, adaptable regions for local exploration, and forbidden regions for clash avoidance. A unified all-atom representation supports topology-preserving adaptation across diverse binder categories, while Pareto selection preserves nondominated candidates across the two objectives. We evaluate FlexEvo across multiple generation baselines and nine representative binder categories spanning diverse molecular sizes and structural topologies. FlexEvo reduces the category-balanced mean relative performance degradation from 47.8% to 4.4%, while adding only 1.4--3.1 minutes of adaptation per sample. These results establish single-state inference-time adaptation as a practical route toward robust biomolecular complex design across protein conformational landscapes.

Computational design of protein conformational landscapes for metamorphic behavior
A. Khramushin, M. Elia, C. Schellhaas +8
bioRxiv13 Oct 2026PreprintCC BYFull textSystems & synthetic biologybioRxiv
SARS-CoV-2 Nucleocapsid Protein forms biomolecular condensates and interacts with ALS-associated RNA binding proteins in vitro
J.-P. W. E. Clarke, A. Keating, V. Noches +4
bioRxiv29 Sep 2026PreprintAll rights reservedFull textNeurosciencebioRxiv
Pocket contrastive learning and conformation fusion for robust structure-conditioned 3D molecular generation under receptor conformational variation.
Huan Liu, Yijie Ding, Xiaoyi Guo +1
Bioinformatics (Oxford, England)1 Oct 2026Peer-reviewedCC BYFull textBioinformatics & computational biologyEurope PMCCrossref
TopoFlow: Evolutionarily Conditioned Flow Matching for Protein Conformational Ensemble Generation
X. Yang, X. Cui, X. LI +4
bioRxiv22 Sep 2026PreprintAll rights reservedFull textBioinformatics & computational biologybioRxiv
Mice navigate complex landscapes by learning a sparse vocabulary of paths
T. Ma, A. E. Comrie, V. Pjanovic +4
bioRxiv10 Oct 2026PreprintCC BYFull textZoology & animal behaviourbioRxiv

Cite

BibTeX RIS

BibTeX

@misc{zeng2026robust,
  title = {{Robust Biomolecular Complex Design Across Protein Conformational Landscapes}},
  author = {Qingyuan Zeng and Zongqi Xu and Anglin Liu and Ziqi Gong and Pengxiang Cai and Zixin Guan and Yunan Chen and Sen Gao and Min Zhou and Jintai Chen},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2609.33726},
  eprint = {2609.33726},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.33726},
  note = {CC BY-NC-SA 4.0},
}

RIS

TY  - UNPB
TI  - Robust Biomolecular Complex Design Across Protein Conformational Landscapes
AU  - Qingyuan Zeng
AU  - Zongqi Xu
AU  - Anglin Liu
AU  - Ziqi Gong
AU  - Pengxiang Cai
AU  - Zixin Guan
AU  - Yunan Chen
AU  - Sen Gao
AU  - Min Zhou
AU  - Jintai Chen
PY  - 2026
DA  - 2026/10/02
PB  - arXiv
DO  - 10.48550/arxiv.2609.33726
UR  - https://arxiv.org/abs/2609.33726
AB  - Proteins populate conformational ensembles, yet structure-based biomolecular design typically optimizes candidates against a single target conformation. Consequently, a candidate that fits one state can lose favorable interactions or develop steric clashes when the target adopts another. We introduce FlexEvo, a model-agnostic evolutionary framework that adapts candidates once at inference time from a single target conformation to improve compatibility with alternative natural conformations unseen during adaptation, without retraining the source model or requiring a conformational ensemble. FlexEvo casts cross-state adaptation as geometry-constrained bi-objective optimization, balancing preservation of input-state interactions against robustness to plausible conformational perturbations. To limit the search space and reduce invalid structural edits, geometry-derived FlexBoxes define protected anchor regions, adaptable regions for local exploration, and forbidden regions for clash avoidance. A unified all-atom representation supports topology-preserving adaptation across diverse binder categories, while Pareto selection preserves nondominated candidates across the two objectives. We evaluate FlexEvo across multiple generation baselines and nine representative binder categories spanning diverse molecular sizes and structural topologies. FlexEvo reduces the category-balanced mean relative performance degradation from 47.8% to 4.4%, while adding only 1.4--3.1 minutes of adaptation per sample. These results establish single-state inference-time adaptation as a practical route toward robust biomolecular complex design across protein conformational landscapes.
ER  -

Abstract

Related

Cite